front-end-checklist
Pass
Audited by Gen Agent Trust Hub on May 2, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Risk. The skill is designed to analyze untrusted front-end source code (HTML, CSS, JS) from the user's environment (File:
SKILL.md). Ingestion points: Local project files via user request. Boundary markers: Absent; no instructions define delimiters or advise the agent to ignore embedded directives. Capability inventory: File-read (references) and file-write (to thedocs/folder) capabilities are present inSKILL.md. Sanitization: None; the skill does not specify validation or filtering of the source content before analysis. - [COMMAND_EXECUTION]: Local File System Access. The skill workflow instructs the agent to "Register the checklist session result in the project's docs/ folder if applicable," which constitutes an autonomous write operation based on potentially untrusted input processing (File:
SKILL.md). - [EXTERNAL_DOWNLOADS]: Reference to External Resources. The reference documentation (
references/*.md) contains links to numerous third-party tools and services. These references target well-known technology organizations and reputable services (e.g., Google Developers, Mozilla MDN, W3C, and OWASP) and are provided as educational resources for the developer rather than automated downloads by the skill.
Audit Metadata